Novel Bispecific Antibodies Bispecific antibodies (BiAb) are a promising class of protein pharmaceuticals utilized to target and/or redirect therapeutic cells. Numerous BiAbs are under development to direct activated T cells to tumors. Our laboratory (Lee, 2006) is utilizing this approach to address a major problem in stem cell (SC) therapy: targeting to and retention of SC at sites of tissue injury, e.g. myocardial infarction, stroke, macular degeneration, etc. Production of BiAbs by hybrid hybridomas or chemical cross- linking suffers from several disadvantages, such as aggregation and possible inactivation (in the case of chemical ligation), inefficiency of production and separation of unwanted byproducts. Attempts to solve these problems by genetic engineering have been largely unsuccessful, raising new problems, including reproducibility and potency. No currently available routine techniques permit commercial scale production of bivalent bispecific antibodies. However, in the mucosal immune system J chain facilitates the facile assembly of large quantities of pentameric IgM and secretory IgA. Based on this observation we have designed a new, efficient strategy called "JLINK" to create BiAbs. J chain is fused to the C terminus of one IgG antibody and mediates the efficient interaction and covalent linkage to a second antibody which has the tailpiece of IgA fused to the C terminus of its heavy chain. JLINK will permit efficient, simple production of BiAbs for novel therapeutics and diagnostics. [unreadable] To demonstrate proof-of principle we will create a JLINK version of Her2Bi, a BiAb being developed by our collaborator, Dr. L. Lum. Her2bi (anti-CD3 x anti-Her2), a chemically conjugated BiAb targeting activated T cells to HER2 expressing tumor cells shows promise in Phase I/II trials versus breast and prostate cancer (Lum, 2004, 2005, 2006). [unreadable] A. Phase I Specific Aims [unreadable] AIM #1: Construct JLINK IgG fusions to CH3a and J-Chain. [unreadable] AIM #2: Express Bispecific Abs and optimize production. [unreadable] AIM #3: Test Bispecific Abs for activity in vitro. [unreadable] After successful creation of genetically engineered Her2Bi, in Phase II we plan to use JLINK to construct several sets of proprietary BiAbs with clinical potential. [unreadable] [unreadable]